[Ocfs2-users] unable to reliably mount two nodes

[Lee Musgrave]

> hi,
> have installed two identical servers with ubuntu 12.04, and installed
> drbd8-utils and ocfs2-tools.
>
> i have configured drbd.conf as:
>


> # You can find an example in  /usr/share/doc/drbd.../drbd.conf.example
>
> include "drbd.d/global_common.conf";
> include "drbd.d/*.res";
>
> resource datacluster {
>         protocol C;
>
>         handlers {
>                 pri-on-incon-degr "echo o > /proc/sysrq-trigger ; halt -f";
>                 pri-lost-after-sb "echo o > /proc/sysrq-trigger ; halt -f";
>                 local-io-error "echo o > /proc/sysrq-trigger ; halt -f";
>                 outdate-peer "/usr/lib/heartbeat/drbd-peer-outdater -t 5";
>         }
>
>         startup {
>                 wfc-timeout 120;
>                 degr-wfc-timeout 120;
>                 become-primary-on both;
>         }
>
>         disk {
>                 on-io-error detach;
>   fencing resource-and-stonith;
>   }
>
>   net {
>   allow-two-primaries;
>   cram-hmac-alg sha1;
>   shared-secret "password";
>   after-sb-0pri discard-zero-changes;
>   after-sb-1pri discard-secondary;
>   after-sb-2pri disconnect;
>   }
>
>   syncer {
>   rate    100M;
>   verify-alg      sha1;
>   al-extents 257;
>   }
>
>   on nas1 {
>   device /dev/drbd0;
>   disk /dev/xvdb2;
>   address 192.168.10.21:7788;
>   meta-disk /dev/xvdb1[0];
>   }
>
>   on nas2 {
>   device /dev/drbd0;
>   disk /dev/xvdb2;
>   address 192.168.10.22:7788;
>   meta-disk /dev/xvdb1[0];
>   }
> }
>
>
> and configured cluster.conf as:
>



> cluster:
>  node_count = 2
>  name = dcl
>


> node:
>  ip_port = 7777
>  ip_address = 192.168.10.21
>  number = 1
>  name = nas1
>  cluster = dcl
>
> node:
>  ip_port = 7777
>  ip_address = 192.168.10.22
>  number = 2
>  name = nas2
>  cluster = dcl
>
>

>
> everything appears to be working fine. drbd-overview shows the device is
> connected as primary on both servers, i've not noticed any problems at any
> time with the drbd configuration.
>
> the drbd0 device was formatted using mkfs.ocfs2 -T mail /dev/drbd0
> again everything seems fines.
> i can mount the device (mount -t ocfs2 /dev/drbd0 /var/clustermount) on
> either server with no problems, unmount it, mount it on the other server,
> unmount it. i can do this repeatedly without errors. and echo "slotmap" |
> debugfs.ocfs2 -n /dev/drdb0 show the slotmap exactly as expected at any
> point.
>
> when i have the cluster mounted on nas1, and try to simultaneously mount
> it on nas2, it takes a long time, causes disk activity (always on the same
> disk) to go through the roof, then i get a pingack not received in time
> message, nas1 is forcibly rebooted, and /dev/drbd0 gets successfully
> mounted on nas2.
> when nas1 is rebooted, i can mount it, and the slotmap shows the cluster
> successfully mounted on both servers.
> if i try mounting to both servers in any of way it fails, eg, mount to
> nas2 first, then mount to nas1, nas2 gets rebooted, it may or may not
> succesfully mount on nas1, if it was successful, trying to mount it on nas2
> after it's reboot will forcibly reboot nas1.
>
> anyone have any idea what could be going on here? cause it's doing my head
> in.
>
> just in case it's relevant, these 2 servers are virtualized servers on
> xenserver 6.2 for testing purposes, once testing is successful the
> configuration will be installed on identical physical servers.
>
> also can anyone confirm, if i get this working properly, if i export the
> ocfs2 cluster to an 20 node apache server farm using nfs, concurrent writes
> to /var/www/vhosts/*/logs/*.log will work ok? or will each apache node need
> to be a part of the cluster configuration?
>
>
> thanks
> lee.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-users/attachments/20130917/7d3e8703/attachment.html